Problems with soil moisture readings **solved**

[rainrainrain]

I'm using a Meteostick to read a Davis Vantage Pro2 and a Leaf/Soil Moisture station, the Pro2 on ID1 (shown as 0 on the Meteohub) and the Soil on ID2 (shown as 1 on the Meteohub). The Meteostick and Meteohub read both of them fine and the Pro2 gets perfect readings, but for the soil readings I'm getting 0cbar on both sensors I'm using:

Code: Select all

Type	#	Last Signal	Sensor Data
wind	0			2s -63db	0.5(0.6)m/s W	
rain	0			7s -62db	0.0mm/h	
indoor	0			59 sec	30.2°C 30% 1017.6mb (1019.4mb) fc:-1	
outdoor	0			9s -65db	23.8°C 52%	
soil 1	1:10			10s -67db	0.0°C 0cbar	
soil 2	1:11			7s -74db	0.0°C 0cbar	
uv	0			1m -63db	3.5uvi	
solar	0			32s -65db	561W/qm

Because of the number of different components involved it's a bit difficult to debug, does anyone have any ideas whether this could be the Davis Soil station, the Meteostick, or the Meteohub? I only have the Davis ISS and not the console since I use the Meteostick to read it, so I can't check things with the console. Meteohub firmware is 5.0c (the latest).

[admin]

The soil/leaf station does send data, but it looks like no temp sensor
is connected and the soil probe is not showing any wetness at all.
To simulate data you might wrap the soil probe into a wet towel.

[rainrainrain]

I followed the Davis instructions about pre-wetting them so they were dripping wet when I "planted" them, which in theory would give a 0cbar reading (or close to it). It's been a few days now and I was expecting it to slowly creep upwards as they dried out, but nothing's changed, both sensors are pegged at 0. I don't have temperature sensors so (according to the Davis manual) it should apply a default compensating temp of 24 degrees, if I can find out what sort of thermistor they use (it looks like a 350K NTC) I might wire one in to see what readings I get.

[admin]

I see, you are already very deeply into this. Connecting a potentiometer instead of the probes
might give you a chance to simulate values. Unfortunately, I don't know which Ohm range is
regarded appropriate. I doubt that short-cutting will do any harm, so no reason why not to
do some experiments with standard resistors. There are very experienced users around
the Vantage in wxforum.net. May be you can ask there what the Ohm values are to give a try.

[rainrainrain]

Making this a separate post because it's not directly related to the previous one: this page has some data from a Davis Soil/Leaf station showing occasional drops to 0cbar, but it doesn't stay there for long. Here's another post from the same author showing a gradual increase in soil moisture, presumably due to the rain shown by the leaf wetness reading.

[rainrainrain]

Thanks, I'm still waiting for my wxforum membership to be approved (after having to retry their captcha process seven times to get in), I just wanted to narrow down the possible causes first, i.e. try and avoid the immediate response of "you should be using a Davis console to get the readings".

In terms of faking out the sensors, Davis use an Irrometer 200SS, which supposedly has a range of 500-30K ohms and is read using a "solid-state AC resistance bridge" according to Irrometer (which you'd normally expect to be done using a DC bridge, so I'm not sure if there's some sort of exotic requirement imposed by the sensors). So I don't know how practical wiring in a standard resistor is going to be, but I'll give it a go if the wxforum folks don't have any suggestions.

[admin]

I checked with my soil/leaf station, where I have one leaf sensor, one soil sensor and one temp sensor connected as you can see in the picture below.

davis.png (260.79 KiB) Viewed 12819 times

This results on an extremely rainy day like today to these Meteobridge live data:

Screen Shot 2014-12-22 at 12.21.36.png (63.06 KiB) Viewed 12819 times

Leaf#15 is pointless, as first leaf sensor port is not connected, but Leaf#16 reading 2nd leaf sensor port reports wetness of 15.
Soil#10 has temp sensor connected to port 1 reporting 10 degrees Celsius.
Soil#11 has soil probe installed to port 2 reporting 7 centibar.

[rainrainrain]

An update, a full week after "planting" the sensors I'm finally getting nonzero readings:

Code: Select all

Type		#	Last Signal	Sensor Data
wind		0	1s -62db		0.5(0.2)m/s NNW	
rain		0	6s -65db		0.0mm/h	
indoor	0	1 min		22.0°C 30% 1022.6mb (1024.4mb) fc:-1	
outdoor	0	9s -63db		18.9°C 79%	
soil 		1	1:10			0s -67db	0.0°C 9cbar	
soil 		2	1:11			14s -68db	0.0°C 5cbar	
uv		0	1s -62db		0.0uvi	
solar		0	12s -64db		0W/qm

That means the readings were (presumably) accurate, it just took an incredibly long time for the sensors to start drying out enough to get a reading (they're in pretty heavy clay soil, presumably that stays wet for a long time, or at least much longer than I thought). I'll keep checking back over the next few days to see what the trend is, but at least after a week it's finally gone nonzero. Now all I need to do is figure out how to get Weather Underground to display the reading...

[admin]

Meteobridge should report the data to WU (just checked the code).
But I don't know how to inspect the data at WU web pages.

[rainrainrain]

Sure, the problem will be in WU. Their track record of updating software and fixing bugs (which extends to just implementing the things they advertise) is pretty bad, so it's quite possible that their claimed ability to report soil moisture readings doesn't actually exist. It's a pity because their forecasting is really good, unfortunately their software development seems to be done by a sole intern who only works on alternate weekends. If anyone knows of an equivalent service (with iOS/Android app support) I'd be interested in hearing about it, I haven't found much so far.

[rainrainrain]

Another update for people who may have run into the same problem: After several more days of prolonged heat and dryness the readings are slowly creeping upwards, 23cbar for the not-so-heavy clay and 11cbar for the heavy clay. If anyone else is having this issue, "planting" the sensors wet as per the Davis instructions into heavy clay soil means you can end up with zero readings for (in my case) up to a week before they start to dry out.

[admin]

Thanks for sharing this experience. We have a more sandy ground here, so values stabilized much faster.

[rainrainrain]

While poking around with the Meteohub GUI I noticed something that may be causing the lack of info on WU: On the "Weather Networks" page in the "Selection of sensors to be used" section there's no options available for the soil wetness/temperature/leaf wetness. The only additional options are "Temp #2" ... "Temp #7". Could this be why WU isn't displaying anything for these sensors?

[admin]

I was somehow on the wrong path, I thought you are talking about Meteobridge.
Looking at Meteohub no soil/leaf data is transported to WU. I will add this with he next update.

[rainrainrain]

OK, thanks! It'd be nice to be able to monitor soil/leaf/etc values as well.